Background The aim of this study was to determine the frequency and type of respiratory complications after adenotonsillectomy in children. A second aim was to assess the ability of preoperative sleep studies to identify children at risk for respiratory complications. Methods Children referred for sleep studies between 1992 and 1998, who underwent adenotonsillectomy within 6 months of the preoperative study, were reviewed. The study focused on two variables: the obstructive apnea and hypopnea index and the oxygen saturation nadir. Medical charts were reviewed for postoperative respiratory complications. Results Three hundred forty-nine children were referred for sleep studies, and 163 met the inclusion criteria. Thirty-four children (21%) had postoperative respiratory complications requiring a medical intervention. Children experiencing respiratory complications were younger (aged < 2 yr; adjusted odds ratio, 4.3; 95% confidence interval, 1.7-11) and had an associated medical condition (odds ratio, 3; 95% confidence interval, 1.4-6.5). A preoperative obstructive apnea and hypopnea index of 5 or more events per hour increased the chance of postoperative respiratory complications (odds ratio, 7.2; 95% confidence interval, 2.7-19.3), as did a preoperative oxygen saturation nadir of 80% or less (odds ratio, 6.4; 95% confidence interval, 2.8-14.5). A preoperative oxygen saturation nadir of 80% or less had a likelihood ratio of 3.1, increasing the probability of postoperative respiratory complications from 20 to 50%. Conclusions The data suggest, but do not prove, that preoperative nocturnal oximetry could be a useful preoperative test to identify children who are at increased risk for postoperative respiratory complications.